← back

Price

—

Travel

—

Access

—

Social

—

The rapid growth in the capabilities of artificial intelligence platforms opens a multitude of questions about the future of engineering processes and practices. Power electronics, which supports a more sustainable energy landscape through the efficient conversion and control of electric power in energy generation and storage, information technologies, industrial processes, and transportation, is no exception. In this talk, rather than attempting to provide definitive answers to any of the questions, challenges, or opportunities brought about by the proliferation of AI, the goal is to provide an academic perspective on the current and potential future impact of AI, based on recent research and educational activities in the Colorado Power Electronics Center (CoPEC) at the University of Colorado Boulder. A distinction is made between interactive uses of “generative AI” and rapidly emerging “agentic AI” systems. In the former, chat-style interactions with an AI platform target specific research or educational tasks. Agentic AI systems, in contrast, can plan, use tools, and execute multistep tasks with a degree of autonomy and are likely to yield more radical changes in engineering practice. Applications of generative AI as a tool are discussed in the context of a CoPEC research project focused on modal analysis of distributed control in a system based on series-stacked AC modules. Next, a toy investigation of a converter synthesis problem is discussed as an example of pushing generative AI into corners of power electronics often considered to require greater creativity or inventiveness. Finally, the potential for academic research to contribute to the development of agentic AI engineering systems is briefly discussed. The talk concludes with a perspective on education, considering what and how we should teach and learn to prepare future power electronics engineers and researchers for a profession affected by AI.

• Talks